dltHub

dltHub is an AI-native data engineering platform designed for teams of humans and coding agents to generate and manage high-quality data at scale. Operating as an agentic data layer on top of any data warehouse, the platform combines an open-source Python library with managed infrastructure, running ingestion, transformation, orchestration, and managed operations on behalf of the user. It allows coding agents such as Claude Code, Cursor, or Codex to interact with a context graph containing lineage, schema, data quality, governance, and run states.

This enables users to build pipelines for custom sources like Oracle, SAP HANA, IBM DB2, and thousands of other REST APIs, databases, or files. Expectations and schema contracts run on incoming data to halt bad records before they enter the warehouse, while a centralised failure logging and healing surface routes diagnostics directly to the responsible owner. dltHub supports data transformation by generating canonical models and keeping dbt projects running on the same scheduler. It features specific capabilities for regulated life sciences and financial services, providing reproducible, deterministic outputs with built-in audit trails, git-tagged validated baselines, and lineage tracking mapped to standards such as GxP, SR 11-7, SOX ITGC, and BCBS 239.

The underlying dlt open-source library is Apache 2.0 licensed and freely available, enabling teams to use it as an open-source ingestion foundation before transitioning to dltHub when managed runtime, observability, and governed collaboration are required. dltHub integrates with major cloud data platforms including Snowflake, Databricks, BigQuery, and Microsoft Fabric.